JMIR Publications

ABSTRACT

Background:

A crucial global public health concern now is addressing mental health problems. Due to their cost-effectiveness and accessibility, conversational agent interventions (CAIs) have promise in the field of mental health care.

Objective:

In order to present a thorough summary of the traits of CAIs now available for a range of mental health problems, we found for evidence of efficacy and analyzed the significant moderators of efficacy via a meta-analysis of randomized controlled trial.

Methods:

Online databases (EMBASE, MEDLINE, PsycINFO, CINAHL, Web of Science, Cochrane databases) were systematically searched dated from the establishment of the database to 30 October 2021 and updated to 1 May 2022. Randomized controlled trials comparing CAIs with any other type of control condition in improving depressive symptoms, generalized anxiety symptoms, specific anxiety symptoms, quality of life/well-being, general distress, stress, mental disorder symptoms, psychosomatic disease symptoms, and positive and negative affect were considered eligible studies. This study followed the PRISMA guidelines. Data were extracted by two independent reviewers, checked by a third, and pooled using both random-effect models or fixed-effects model. Hedges’ g was chosen as the effect size.

Results:

Of 6900 identified records, a total of 32 studies were included, involving 6089 participants. CAIs showed significant short-term effects compared with control conditions in improving depressive symptoms (g = 0.29; 95% CI, 0.20 to 0.38); generalized anxiety symptoms (g = 0.29; 95% CI, 0.21 to 0.36); specific anxiety symptoms (g = 0.47; 95% CI, 0.07 to 0.86); quality of life/well-being (g = 0.27; 95% CI, 0.16 to 0.39); general distress (g = 0.33; 95% CI, 0.20 to 0.45); stress (g = 0.24; 95% CI, 0.08 to 0.41); mental disorder symptoms (g = 0.36; 95% CI, 0.17 to 0.54); psychosomatic disease symptoms(g = 0.62; 95% CI, 0.14 to 1.11); and negative affect (g = 0.28; 95% CI, 0.05 to 0.51). However, the long-term effects of CAIs for the most mental health outcomes were not significant (g = -0.04 to g = 0.39). Personalization and empathic response were two critical facilitators of efficacy. The longer duration of interaction with conversational agents was associated with the larger pooled effect sizes.

Conclusions:

Findings show that CAIs are research-proven interventions that ought to be implemented more widely in mental health care. CAIs are effective and easily acceptable for those with mental health problems. The clinical application of this novel digital technology will conserve human health resources and optimize the allocation of mental health services.